Invention Publication
- Patent Title: System and Method for Pose Tolerant Feature Extraction Using Generated Pose-Altered Images
-
Application No.: US18011558Application Date: 2022-03-28
-
Publication No.: US20240013457A1Publication Date: 2024-01-11
- Inventor: Marios SAVVIDES , Kai Hu , Yutong Zheng , Ahmed Uzair , Sreena Nallamothu
- Applicant: Marios SAVVIDES , Kai Hu , Yutong Zheng , Ahmed Uzair , Sreena Nallamothu , CARNEGIE MELLON UNIVERSITY
- Applicant Address: US PA Wexford
- Assignee: Marios SAVVIDES,Kai Hu,Yutong Zheng,Ahmed Uzair,Sreena Nallamothu,CARNEGIE MELLON UNIVERSITY
- Current Assignee: Marios SAVVIDES,Kai Hu,Yutong Zheng,Ahmed Uzair,Sreena Nallamothu,CARNEGIE MELLON UNIVERSITY
- Current Assignee Address: US PA Wexford
- International Application: PCT/US22/22112 2022.03.28
- Date entered country: 2022-12-20
- Main IPC: G06T11/60
- IPC: G06T11/60 ; G06T17/00 ; G06T3/60 ; G06V10/44 ; G06V10/56 ; G06V10/74 ; G06V20/68

Abstract:
Disclosed herein is a system and method for augmenting data by generating a plurality of pose-altered images of an item from one or more 2D images of the item and using the augmented data to train a train a feature extractor. In other aspects of the invention, the trained feature extractor is used to enroll features extracted from images of new products in a library database of known products or to identify images of unknown products by matching features of an image of the unknown product with features stored in the library database.
Public/Granted literature
- US11900516B2 System and method for pose tolerant feature extraction using generated pose-altered images Public/Granted day:2024-02-13
Information query